Jim Hoffman was asking about ways to get transcriptions of audio lectures on-line via voice->text conversion.
No, off-hand I do not know any faculty who are doing this; however, that doesn’t mean no one is doing it (including students). Nor do I know of an on-campus resource that can offer much support (The Center for Multimedia Development in the library doesn’t have the software, and Roger or Judi would probably send you to me if you asked them about it. I haven’t heard of the Center for Teaching and Learning doing anything like this, but they might).
However, after poking around the ScanSoft web site, it looks like (in theory), what you propose is pretty straight-forward, with the addition of a digital voice recorder or PDA and the right version of Dragon NaturallySpeaking.
The "right version of Dragon NaturallySpeaking" appears to be version 7 or 8 of the "Preferred" or "Professional" product (not "standard").
ScanSoft has certified a number of handheld devices, meaning "Passed all tests with Dragon NaturallySpeaking and meets our highest performance and quality standards." In particular, the Sony ICD-ST25 Digital Voice Recorder, which can be found for around $150. About $50 more bundled with Dragon Naturally Speaking Preferred Voice-to-Print Software Version 7.0.
In any case, you record to your handheld device, upload sound via USB, Dragon NaturallySpeaking does the rest.